About Opentopia
Opentopia Directory Encyclopedia Tools

Chroma subsampling

Encyclopedia : C : CH : CHR : Chroma subsampling

Chroma subsampling is the use of implementing more resolution for the (quantity representative of) luminance information than the (quantity representative of) color information. It is used in many video encoding schemes (both analog and digital) and also in JPEG encoding.

Contents

Why subsampling works

Because the human eye is less sensitive to color than luminance, bandwidth can be optimized by storing more luminance detail than color detail. At normal viewing distances, there is no perceptible loss incurred by sampling the color detail at a lower rate. In video systems, this is achieved through the use of color difference components. The signal is divided into a luma (Y') component and two color difference components (chroma).

Chroma sumsampling deviates from color science in that the luma and chroma components are formed as a weighted sum of gamma-corrected (tristimulus) R'G'B' components instead of linear (tristimulus) RGB components. As a result, luminance and color detail are not completely independent of one another. There is some "bleeding" of luminance and color information between the luma and chroma components. The error is greatest for highly-saturated colors and can be somehwat noticeable in between the magenta and green bars of a color bars test pattern (that has chroma sumsampling applied). This engineering approximation allows color subsampling to be more easily implemented.

Sampling systems and ratios

The subsampling scheme is commonly expressed as a three part ratio (i.e. 4:2:2), although sometimes expressed as four parts (i.e. 4:2:2:4). The parts are (in their respective order): Chroma subsampling ratios.png

The mapping examples given are only theoretical and for illustration. The bitstreams of real-life implementations will most likely differ. Also note that the diagram does not show chroma filtering or from which pixels the chroma values are sampled from.

Why is this done?

Video usually requires massive amounts of bandwidth. Because of storage and transmission limitations, there is always a desire to reduce (or compress) the signal. 4:4:4 video, whether RGB or Y'CbCr, is considered lossless. Since the human visual system is much more senstive to variations in brightness than colour, we can take Y'CbCr and reduce the Cb and Cr components to half the bit resolution, giving 4:2:2. The end result is a reduced bandwdith video signal with little impact on what is perceived by the viewer.

Types of subsampling

4:4:4 Y'CbCr

Each of the three channels has the same sample rate, so each single pixel in the resulting image gets three full words (usually 8 or 10 bits long) of information, resulting in 3 bytes per pixel for 8-bit quantization when not using compression.

Mapping:

The bitstream 

Y0 Cb0 Cr0 Y1 Cb1 Cr1 Y2 Cb2 Cr2 Y3 Cb3 Cr3
will map to the following four pixels: 
[Y0 Cb0 Cr0] [Y1 Cb1 Cr1] [Y2 Cb2 Cr2] [Y3 Cb3 Cr3]
This is the best color sampling ratio (it yields a nearly perfect representation of each pixel's color), and is used as an intermediate format in high-end film scanners and cinematic postproduction.

4:4:4 R'G'B' (no subsampling)

Note that "4:4:4" may instead be referring to R'G'B' color space, which implicitly does not have any chroma subsampling at all. Formats such as HDCAM SR can record 4:4:4 R'G'B' over dual-link HD-SDI.

4:2:2

Each of the two color-difference channels has half the sample rate of the luma channel, so horizontal color resolution is only half that of 4:4:4. For uncompressed video and 8-bit quantization, each macropixel of two neighbouring pixels uses 4 bytes of memory.

Mapping:

The bitstream 

Y0 Cb0 Y1 Cr1 Y2 Cb2 Y3 Cr3
will map to the following four pixels: 
[Y0 Cb0 Cr1] [Y1 Cb0 Cr1] [Y2 Cb2 Cr3] [Y3 Cb2 Cr3]
This is still a very good quality, and most higher-end digital video formats use this ratio:

4:2:1

Although this mode is technically defined, very few software or hardware codecs use this sampling mode.

4:1:1

In 4:1:1 chroma subsampling, the horizontal color resolution is quartered. This is still acceptable for lower-end and consumer applications. Uncompressed video in this format with 8-bit quantization uses 6 bytes for every macropixel (4 pixels in a row).

Mapping:

The bitstream 

Y0 Cb0 Y1 Y2 V2 Y3
will map to the following four pixels: 
[Y0 Cb0 V2] [Y1 Cb0 V2] [Y2 Cb0 V2] [Y3 Cb0 V2]
Formats that use 4:1:1 chroma subsampling include:

4:2:0

This scheme is found in:

Cb and Cr are each subsampled at a factor of 2 both horizontally and vertically. Cb and Cr are effectively centered vertically halfway between image rows.

There are three variants of 4:2:0 schemes, having different horizontal and vertical siting.

The PAL and SECAM color systems are especially well-suited to this kind of data reduction. Most digital video formats corresponding to PAL use 4:2:0 chroma subsampling, with the exception of DVCPRO25 (it uses 4:1:1 chroma subsampling). Compressed video in this format with 8-bit quantization uses 6 bytes for every macropixel (2×2 pixels).

Mapping:

The bitstream 

Yo0 Cbo0 Yo1 Yo2 Cbo2 Yo3

Ye0 Ve0 Ye1 Ye2 Ve2 Ye3
will map to the following two lines of four pixels each: 
[Yo0 Cbo0 Ve0] [Yo1 Cbo0 Ve0] [Yo2 Cbo2 Ve2] [Yo3 Cbo2 Ve2]

[Ye0 Cbo0 Ve0] [Ye1 Cbo0 Ve0] [Ye2 Cbo2 Ve2] [Ye3 Cbo2 Ve2]

4:1:0

This ratio is possible (indeed, some codecs do support it), but not widely used, since its color fidelity is even below that of VHS. It means half the vertical and quarter the horizontal color resolutions, with only one eighth of the bandwidth of the maximum color resolutions used. Uncompressed video in this format with 8-bit quantization uses 10 bytes for every macropixel (4 x 2 pixels).

Mapping:

The bitstream 

Yo0 Cbo0 Yo1 Yo2 Yo3

Ye0 Ve0 Ye1 Ye2 Ye3
will map to the following two lines of four pixels each: 
[Yo0 Cbo0 Ve0] [Yo1 Cbo0 Ve0] [Yo2 Cbo0 Ve0] [Yo3 Cbo0 Ve0]

[Ye0 Cbo0 Ve0] [Ye1 Cbo0 Ve0] [Ye2 Cbo0 Ve0] [Ye3 Cbo0 Ve0]

Terminology

The term Y'UV refers to an analog encoding scheme while Y'CbCr refers to a digital encoding scheme. One difference between the two is that the scale factors on the chroma components (U, V, Cb, and Cr) are different. However, the term YUV is often (erroneously) used to refer to Y'CbCr encoding. Hence, terms like "4:2:2 YUV" likely refer to 4:2:2 Y'CbCr. The exact meaning may be ambiguous and needs to be derived from context.

In a similar vein, the term luminance and symbol Y is often (erroneously) used to refer to luma, denoted with the symbol Y'. Note that the luma (Y') of video engineering deviates from the luminance (Y) of color science (as defined by CIE). Luma is formed as the weighted sum of gamma-corrected (tristimulus) RGB components. Luminance is formed as a weighed sum of linear (tristimulus) RGB components.

In practice, the CIE symbol Y is often incorrectly used to denote luma. In 1993, SMPTE adopted Engineering Guideline EG 28, clarifying the two terms. Note that the prime symbol ' is used to indicate gamma correction.

Similarly, the chroma/chrominance of video engineering differs from the chrominance of color science. The chroma/chrominance of video engineering is formed from weighted tristimulus components, not linear components. In video engineering practice, the terms chroma, chrominance, and saturation are often (and perhaps ambiguously!) used to refer to the same concept.

References

Poynton, Charles. "YUV and luminance considered harmful: A plea for precise terminology in video" [link] Poynton, Charles. "Digital Video and HDTV: Algorithms and Interfaces." USA: Morgan Kaufmann Publishers, 2003.

Suggested Reading

Poynton, Charles. "Digital Video and HDTV: Algorithms and Interfaces." USA: Morgan Kaufmann Publishers, 2003.

 

From Wikipedia, the Free Encyclopedia. Original article here. Support Wikipedia by contributing or donating.
All text is available under the terms of the GNU Free Documentation License See Wikipedia Copyrights for details.

Search Titles
0123456789
ABCDEFGHIJ
KLMNOPQRST
UVWXYZ?

E-mail this article to:

Personal Message: